This bundle contains datasets considered to evaluate CoNeRa. In particular, the bundle includes:
1) for each dataset, 5 training files and 5 testing files constructed by performing the 5-fold cross validation of the original node set (arff format);
2) for each training file, 5 different initial portions of labeled node set are sampled (txt format);
3) for each dataset, the edge structure of the network  (csv format);
4) the network configuration file (nh_settings.ini) that reports the neighbourhood size, the minimum and maximum step used to construct the neighbourhood, the power of the weight;
5) the active learning configuration file (STDDEV.ini) that reports percentage of training examples to be labeled per iteration, the percentage of examples to be labelled before stopping the active learning process, the disagreement measure for the active example selection, the base regression learner, the collective inference schema with the discretization algorithm (if required).


